]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: c0d0430)
llama : use output_resolve_row() in get_logits_ith/get_embeddings_ith (#19663)
authorDaniel Bevenius <redacted>
Thu, 19 Feb 2026 08:48:08 +0000 (09:48 +0100)
committerGitHub <redacted>
Thu, 19 Feb 2026 08:48:08 +0000 (09:48 +0100)
commiteacb4b67a2905d1f53e5f55033f81c4824b49e64
tree7c4e3d4b219b7a950392990d779e3eb07fa8a62dtree
parentc0d04303400e64a798506e3f2342940ae268db15commit | diff
llama : use output_resolve_row() in get_logits_ith/get_embeddings_ith (#19663)

This commit updates get_logits_ith(), and get_embeddings_ith() to use
output_resolve_row() to resolve the batch index to output row index.

The motivation for this is to remove some code duplication between these
functions.
src/llama-context.cpp diff | blob | history
Packaging of ggml-org/llama.cpp